OmniVoice is a massively multilingual zero-shot text-to-speech (TTS) model supporting over 600 languages. Built on a novel diffusion language model-style architecture, it delivers high-quality speech with superior inference speed, supporting voice cloning and voice design.
Underdog-Woof
ConwayResearch·released 5 Nov 2026
hf download ConwayResearch/Underdog-WoofUnderdog Woof is a small English text-to-speech model with two voices: Juno and Mira. It reads phone numbers, dates, prices and codes the way people say them, and it runs on a laptop CPU with nothing but numpy, from a single 40.6 MB file. Try it in your browser at underdog.ai/woof.
How much VRAM does Underdog-Woof need?
The parameter count of this model is unknown, so there is no memory estimate yet.
Run Underdog-Woof locally
No runtime support recorded yet. Most models also run through Hugging Face Transformers.
Quantised versions and downloads
| Repository | Format | Quant | Size |
|---|---|---|---|
| cowWhySo/Underdog-Woof-MLX | MLX | — | — |
| masterofaudio2077/underdog-woof-onnx | ONNX | — | — |
About Underdog-Woof
Underdog Woof is a small English text-to-speech model with two voices: Juno and Mira. It reads phone numbers, dates, prices and codes the way people say them, and it runs on a laptop CPU with nothing but numpy, from a single 40.6 MB file. Try it in your browser at underdog.ai/woof. Each voice reads "Please call me back at 415-555-0199 before 5:30 p.m. on Friday." (seed 1). More clips are in the samples folder, listed below. Woof comes in two sizes. This repo is the two-voice model. The one-voice model is the one that runs in your browser at underdog.ai/woof. The one-voice speed is plain PyTorch with its int8 weights at the default 16 steps; the two-voice speed is this repo's numpy runtime. Fish Audio Drama 3, for scale, is a 41.5 GB download and gets 1.89% wrong. Word error rate on Gradium TTS Hard Cases (English, 500 sentences), every model run and scored by us the same way, each in its best-scoring voice. Lower is better. Kitten Nano 0.8 is the only smaller download, and it gets 9.81% of words wrong: nothing we tested is both smaller and more accurate than Woof. Transcripts by qwen3.8-omni-flash, word errors counted by gemma-4-31b-it. Download is the smallest official set of weights each model needs to run, from its Hugging Face repo in November 2026 (Phonon 31m: the 4-bit GGUF; Kitten Nano: the int8 ONNX plus voices; Pocket TTS: the English model; Qwen3-TTS: the model plus its speech tokenizer). Cloud-only models don't publish a size.
Full model card on Hugging Face →Alternatives to Underdog-Woof
All alternatives →VoxCPM2 is a tokenizer-free, diffusion autoregressive Text-to-Speech model — 2B parameters, 30 languages, 48kHz audio output, trained on over 2 million hours of multilingual speech data.
Qwen3-TTS covers 10 major languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, and Italian) as well as multiple dialectal voice profiles to meet global application needs.
Qwen3-TTS is a family of advanced multilingual, controllable, robust, and streaming text-to-speech models. Trained on over 5 million hours of speech data spanning 10 languages, Qwen3-TTS supports state-of-the-art 3-second voice cloning and description-based control.
Kokoro is an open-weight TTS model with 82 million parameters. Despite its lightweight architecture, it delivers comparable quality to larger models while being significantly faster and more cost-efficient.
Qwen3-TTS is a series of advanced multilingual, controllable, robust, and streaming text-to-speech models developed by the Qwen team.
Underdog-Woof FAQ
- Can I use Underdog-Woof commercially?
- No. It is released under underdog-non-commercial-1.0. Custom licence: read the model card before using it commercially.
Rate this model
Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.
0 comments